Exercise induces metabolic changes including loss of weight, reduction in triglyceride and LDL levels, increased HDL, and enhanced insulin sensitivity. Moderate physical activity has been shown to have beneficial effect in preventing heart disease, heart attack and stroke, and has been implicated as a preventive measure for some cancers. Genetic predisposition to complex traits, such as propensity for voluntary exercise, results from effects of combinations of genetic variations within a number of polygenes, known as quantitative trait loci (QTL). Remarkably, no QTL have been reported for voluntary exercise in any species. In three specific aims, our goal is to address this gap in knowledge by identifying QTL controlling voluntary exercise, and its underlying transcriptional landscape, using a unique polygenic mouse model: Aim 1: Localize QTL for voluntary exercise: We will create an F4 population of 800 mice originating from a cross between the HR line, selectively bred for increased voluntary exercise, and C57BI/6J (B6). This population will be phenotyped for voluntary exercise, caloric intake, and weight gain and body composition. All mice will be genotyped for 768 evenly spaced and fully-informative SNP markers. Aim 2: Localize expression QTL (eQTL) for voluntary exercise: We will evaluate global gene expression in hippocampus and gastrocnemius muscle from HR x B6 F4 mice with whole-genome microarrays. Using genotypes from Aim 1, we will map eQTL that regulate significant variation in mRNA abundance. Aim 3: Identify candidate genes underlying QTL for voluntary exercise: Merging results from Aims 1 and 2, we will identify subsets of cis-acting eQTL that map under peaks of QTL for voluntary exercise, and for which the underlying expression phenotypes are correlated with voluntary exercise. These will be subjected to classical cis-trans tests to directly validate the presence of cis-acting regulatory variation within candidate genes for voluntary exercise predisposition. These synergistic and integrated methods, applied to a powerful polygenic mouse model that has undergone long-term selective breeding for increased running, offer a unique opportunity to evaluate the genetic architecture of predisposition to voluntary exercise. This research will discover QTL for voluntary exercise, a poorly understood yet important component of obesity and overall health, and advance to identify their underlying genes.